Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 717
    Prev
    Next
    INDEX
    Explanations

    legal and political terms or discussions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    561
    +0.10
    0.4%
    736
    +0.09
    0.4%
    757
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.11
    289
    +0.09
    0.01
    325
    +0.09
    0.06
    Negative Logits
    CVE
    -0.90
    olute
    -0.70
    sburg
    -0.70
    ILY
    -0.69
    nesota
    -0.68
    sylvania
    -0.68
    etics
    -0.68
    icably
    -0.67
    ndum
    -0.67
    ilton
    -0.66
    POSITIVE LOGITS
    backer
    1.16
    ups
    1.05
    breaking
    1.03
    breakers
    0.94
    ages
    0.92
     crossings
    0.91
    aments
    0.90
    breaks
    0.89
    up
    0.87
    break
    0.86
    Activations Density 2.019%

    No Known Activations